1 / 7

Comments on FutureGrid

Comments on FutureGrid. Geoffrey Fox Indiana University 15 November 2011. What is FutureGrid?. The FutureGrid project mission is to enable experimental work that advances: Innovation and scientific understanding of distributed computing and parallel computing paradigms ,

zoltin
Download Presentation

Comments on FutureGrid

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Comments on FutureGrid Geoffrey Fox Indiana University 15 November 2011

  2. What is FutureGrid? • The FutureGrid project mission is to enable experimental work that advances: • Innovation and scientific understanding of distributed computing and parallel computing paradigms, • The engineering science of middleware that enables these paradigms, • The use and drivers of these paradigms by important applications, and, • The education of a new generation of students and workforce on the use of these paradigms and their applications. • The implementation of mission includes • Distributed flexible hardware with supported use • Identified IaaS and PaaS “core” software with supported use • Expect growing list of software from FG partners and users • Outreach

  3. Backdrop • NSF sent 15 months funding for FutureGrid August 12 2011 covering • October 1 2010 to December 31 2011 • PY2 contracts with increased effort in support (finally) in place • Set up skeleton user support to be filled in by hires in next 3 weeks • Substantial discussions on relation to XSEDE • Project submission through XSEDE planned • Filed contribution to XSEDE October 1 quarterly • Day long “All Hands” Software meeting September 22 • When new hires in place , will have • All Hands User Support meeting • Internal Tutorial on systems and software including IaaS and PaaS

  4. Focus on the community impact around the experimental computer science and technology evaluation/exploration facility uniquely provided by this project • Reuse of the hardware money to support additional user support and outreach to the computer science community – defined both by the project management and UAB as critical for maximizing FG’s impact. • NSF and IU approval obtained (it was match money); funding released; setting up letters and interviews with applicants • Two jobs – consultant and outreach • Ensure effort allocated to partners within the project is delivered and there are mechanisms for demonstrating it (e.g. local site system administration effort, software development effort, etc.) • Software tracked in Jira • Will add support effort tracking to Jira • Statements of work and PEP define activities. Delay in PY2 funding delayed some activities • Leverage other dissemination routes (e.g. technology providers, software developers) to communicate with new communities through their peers • Tutorials, Papers etc. at meetings • Project request through XSEDE portal • 76 people ran Nimbus on FutureGrid at Sunday tutorial ; Expt CS workshop Friday at SC11 • Put systems in place to ensure feedback from the current user community is tracked and can be seen to be resolved (transparency & impact) • User survey designed; will be released end of month • Better monitoring of tickets – new hires will help • Make viewing of ticket system open

  5. Determine the prioritized objectives of the project • Ensure the project plan balances the resources in the project to these objectives • Redirecting resources to support of users and popular software platforms • Establish metrics around these objectives so that their success and impact can be tracked • Track use of software systems from IaaS to Paas (Nimbus, Eucalyptus, Openstack, MapReduce, Saga, Genesis, Globus) • Demonstrate that the project resources (e.g. staff) that are being expended by partners demonstrate a return towards the project objectives • Jira and PEP (WBS) • Define who you need to get on board as users to be successful (e.g. measure of success around papers?) and provide a strategy to seek out and attract these users • We have online project results and will align to XSEDE metrics • User support will proactively get results including papers published • Current outreach strategy is talks, tutorials • Focus on the experimental CS aspects on FG resources and collaborate with XSEDE on the support of application science • PY2: 118 non system projects approved, 9 EOT, 3 Interoperability, 15 non life science app, 21 Life Science, 56 CS, 32 Tech evaluation (multiple tags) • The requirements around the use of FG by XSEDE and XSEDE by FG need to be defined by both sides and it is encouraging to see these discussions have started • Work continuing; privileged access process for FutureGrid defined

  6. Enhancements in software packages supported by FG need to demonstrate that there is an impact from the investment made • Define metrics to show that this investment brings in new users (e.g. project activities enabled by the enhanced software environment) • Largest single investment is Nimbus; its use reflects improvement • Show that the software engineering activity is linked back to user requests so as to avoid image that the software development is not FG user focused • Day long software meeting in September to refocus plans • Naturally software effort (dollars) moving to support as project “matures”

  7. It is not evident that the project management has a public project plan that they are implementing and tracking • Currently PEP and biweeklies are publicly available • Will make tickets public • Add support as well as software to Jira • Ensure allocated staff resources are committed by the sites to the objectives agreed by the project and defined in the WBS • WBS in PEP; SoW’s and biweeklies aligned to PEP & JIra • Develop a clear revised WBS that has an allocation of resources against the objectives of the project • PY3 PEP • Ensure there is transparency around the allocation and availability of the resources • Not certain what isn’t transparent • Projects and results open

More Related